Skip to main page content
U.S. flag

An official website of the United States government

Dot gov

The .gov means it’s official.
Federal government websites often end in .gov or .mil. Before sharing sensitive information, make sure you’re on a federal government site.

Https

The site is secure.
The https:// ensures that you are connecting to the official website and that any information you provide is encrypted and transmitted securely.

Access keys NCBI Homepage MyNCBI Homepage Main Content Main Navigation
. 2021 Oct 1:15:733501.
doi: 10.3389/fnins.2021.733501. eCollection 2021.

Transcriptomic Signatures Associated With Regional Cortical Thickness Changes in Parkinson's Disease

Arlin Keo  1   2 , Oleh Dzyubachyk  3 , Jeroen van der Grond  3 , Jacobus J van Hilten  4 , Marcel J T Reinders  1   2   5 , Ahmed Mahfouz  1   2   5
Affiliations

Transcriptomic Signatures Associated With Regional Cortical Thickness Changes in Parkinson's Disease

Arlin Keo et al. Front Neurosci. .

Abstract

Cortical atrophy is a common manifestation in Parkinson's disease (PD), particularly in advanced stages of the disease. To elucidate the molecular underpinnings of cortical thickness changes in PD, we performed an integrated analysis of brain-wide healthy transcriptomic data from the Allen Human Brain Atlas and patterns of cortical thickness based on T1-weighted anatomical MRI data of 149 PD patients and 369 controls. For this purpose, we used partial least squares regression to identify gene expression patterns correlated with cortical thickness changes. In addition, we identified gene expression patterns underlying the relationship between cortical thickness and clinical domains of PD. Our results show that genes whose expression in the healthy brain is associated with cortical thickness changes in PD are enriched in biological pathways related to sumoylation, regulation of mitotic cell cycle, mitochondrial translation, DNA damage responses, and ER-Golgi traffic. The associated pathways were highly related to each other and all belong to cellular maintenance mechanisms. The expression of genes within most pathways was negatively correlated with cortical thickness changes, showing higher expression in regions associated with decreased cortical thickness (atrophy). On the other hand, sumoylation pathways were positively correlated with cortical thickness changes, showing higher expression in regions with increased cortical thickness (hypertrophy). Our findings suggest that alterations in the balanced interplay of these mechanisms play a role in changes of cortical thickness in PD and possibly influence motor and cognitive functions.

Keywords: cortical thickness; gene expression analysis; imaging-genetics; neurodegenerative diseases; neuroimaging data.

PubMed Disclaimer

Conflict of interest statement

The authors declare that this study received funding from AbbVie, Hoffmann-La-Roche, and Lundbeck. The funders were not involved in the study design, collection, analysis, interpretation of data, the writing of this article or the decision to submit it for publication.

Figures

FIGURE 1
FIGURE 1
Principal of partial least squares regression (PLS). (A) Principal component analysis (PCA) and PLS project measurements to a new latent space. Unlike PCA, PLS tries to find a latent space that is maximally correlated with another measurement y from dataset Y on the same samples. (B) The first latent component t1 of dataset X is maximally correlated with the first latent factor u1 of dataset Y. T and U scores determine the outer relations of individual datasets in the model. The coefficient β determines the inner relation between both datasets X and Y in the model (more details in Supplementary Methods). (C) In PLS model-1, we used regional gene expression as input to predict the regional t-statistics of ΔCT. Given the PLS model, R in Eqs. 5 and 6 in Supplementary Methods is used as gene weights. In PLS model-2, we used the same input to predict the t-statistics of correlation coefficients β1 of clinical features from Eq. 1.
FIGURE 2
FIGURE 2
t-Statistics of cortical thickness changes (ΔCT) across cortical regions. (A) CT was assessed between PD patients and healthy controls. Higher t-statistics (red) indicate a larger CT in controls compared to the CT in patients and thus corresponds to cortical atrophy. (B) CT in the left hemisphere compared to the right hemisphere in PD patients. Higher t-statistics (red) indicate a larger CT in the right hemisphere compared to the left hemisphere and thus corresponds to cortical atrophy in the left hemisphere. P-values are BH-corrected and significant regions (P < 0.05) are labeled.
FIGURE 3
FIGURE 3
Transcriptional signatures to predict t-statistics of ΔCT between PD patients and controls in PLS model-1. (A) PLS component-1 scores of predictor variables (gene expression) visualized in cortical regions (lateral and medial view of the left hemisphere). (B) Regression fit of the latent predictor variable, PLS component-1 scores, with the single response variable, CT changes in PD measured as the t-statistics of ΔCT between PD patients (149) and controls (369) across the 34 cortical regions. (C) Mean expression of genes in the top 30 significant pathways (rows) across cortical regions (columns). A complete heatmap with all significant pathways is given in Supplementary Figure 4. The correlation between transcriptomic signatures and CT changes in PD across cortical regions is predicted by the gene weights for PLS component-1 shown in boxplots for each pathway where the median weight is either negative or positive. Negatively correlated pathways show high gene expression in regions with low t-statistics of ΔCT and gene expression decreases in regions with higher t-statistics of ΔCT. In our analysis, negative t-statistics correspond to increased CT (cortical hypertrophy) and positive t-statistics of ΔCT correspond to decreased CT (cortical atrophy). Positively correlated pathways show low expression in regions with low t-statistics of ΔCT and expression increases in regions with higher t-statistics of ΔCT. * = APC:Cdc20 mediated degradation of cell cycle proteins prior to satisfaction of the cell cycle checkpoint. ** = APC/C:Cdh1 mediated degradation of Cdc20 and other APC/C:Cdh1 targeted proteins in late mitosis/early G1.
FIGURE 4
FIGURE 4
Relationship between clinical scores and CT across PD patients. Linear regression was used to predict clinical scores from CT across at most 123 PD patients (Supplementary Figure 1). Separate models were used for each clinical feature (row) and cortical region (column) to obtain regression coefficients, see Eq. 1. The heatmap shows the two-sided t-statistics of the regression coefficient when tested for H0: β1 = 0. Regions (columns) are clustered based on complete linkage of the Euclidean distance of the t-statistics of β1.
FIGURE 5
FIGURE 5
Transcriptional signatures of PLS component-1 in PLS model-2 predictive of the relationship between cortical thickness (CT) and clinical scores. (A) PLS scores for PLS component-1 of the predictor variables (gene expression) and (B) its correlation with PLS component-1 of the response variables (t-statistic of β1 in Eq. 1). Axes show the percentage of explained variance for each component; r indicates the Pearson correlation. (C) Mean expression across cortical regions (columns) of genes in the top 30 significant pathways (rows). A complete heatmap with all significant pathways is given in Supplementary Figure 8. * = Respiratory electron transport, ATP synthesis by chemiosmotic coupling, and heat production by uncoupling proteins.
FIGURE 6
FIGURE 6
Transcriptional signatures of PLS component-2 in PLS model-2 predictive of the relationship between cortical thickness (CT) and clinical scores. (A) PLS scores for PLS component-2 of the predictor variables (gene expression) and (B) its correlation with PLS component-1 of the response variables (t-statistic of β1 in Eq. 1). Axes show the percentage of explained variance for each component; r indicates the Pearson correlation. (C) Mean expression across cortical regions (columns) of genes in the top 30 significant pathways (rows). A complete heatmap with all significant pathways is given in Supplementary Figure 9.
FIGURE 7
FIGURE 7
Correlations between PLS model-2 component-1 and component-2 scores of the predictor variables and individual response variables. Each plot shows the correlation between the predictor variables of gene expression (x-axis) and the response variables which are the relationships between CT and scores of a clinical feature across cortical regions (y-axis). On top of each plot, the Pearson correlation and the Y-loadings (Q in Eqs. 4 and 8) are shown; both values tell something about the sign (−/+) and magnitude (high/low) of the correlation. Each point or sample is one of the 34 cortical regions. Regions are labeled for those with minimum or maximum value along one of the axes.
FIGURE 8
FIGURE 8
Schematic overview of the balance between biological pathways and their influence on CT across cortical brain regions. The big arrow indicates the caudal-to-rostral (red-to-blue) or rostral-to-caudal (blue-to-red) change in CT across cortical brain regions of PD patients with red indicating decreased CT (atrophy) in caudal regions and blue indicating increased CT (hypertrophy) in rostral regions. Genes within pathways associated with sumoylation showed that the expression of these genes within the pathways increases from rostral to caudal regions. Other biological pathways that were correlated with CT changes in PD included regulation of mitotic cell cycle, mitochondrial translation, DNA damage responses, and ER-Golgi traffic, and the involved genes showed decreasing expression patterns from rostral to caudal regions (or increasing from caudal to rostral regions). All enriched pathways shared many common genes and were generally associated with cellular maintenance mechanisms. Literature studies suggest that these biological pathways may be involved in the pathobiology of PD through their interaction with genetic risk variants.
See this image and copyright information in PMC

Similar articles

See all similar articles

Cited by

References

    1. Aarsland D., Creese B., Politis M., Chaudhuri K. R., Ffytche D. H., Weintraub D., et al. (2017). Cognitive decline in Parkinson disease. Nat. Rev. Neurol. 13 217–231. 10.1038/nrneurol.2017.27 - DOI - PMC - PubMed
    1. Altmann-schneider I., de Craen A. J. M., Slagboom P. E., Westendorp R. G. J., van Buchem A., Maier A. B., et al. (2012). Brain tissue volumes in familial longevity: the Leiden longevity study. Aging Cell 11 933–939. 10.1111/j.1474-9726.2012.00868.x - DOI - PubMed
    1. Arnatkevičiūtė A., Fulcher B. D., Fornito A. (2019). A practical guide to linking brain-wide gene expression and neuroimaging data. Neuroimage 189 353–367. 10.1016/j.neuroimage.2019.01.011 - DOI - PubMed
    1. Beatman E. L., Massey A., Shives K. D., Burrack K. S., Chamanian M., Morrison T. E. (2016). Alpha-synuclein expression restricts RNA viral infections in the brain. J. Virol. 90 2767–2782. 10.1128/JVI.02949-15.Editor - DOI - PMC - PubMed
    1. Bologna S., Ferrari S. (2013). It takes two to tango: ubiquitin and SUMO in the DNA damage response. Front. Genet. 4:106. 10.3389/fgene.2013.00106 - DOI - PMC - PubMed

LinkOut - more resources